Annu. Rev. Astron. Astrophys. 1994. 32: 115-52
Copyright © 1994 by . All rights reserved

Next Contents Previous

Neutral Hydrogen Mass and Content

Because of the sensitivity available at centimeter wavelengths, the 21 cm line has been a valuable tool in measuring the redshifts of galaxies and serves as a general indicator of HI content. Since HI line fluxes, or upper limits, are available for some 15000 galaxies skywide, the HI content surpasses nearly all other quantitative indicators of the potential for star formation. The HI content of galaxies has been the subject of recent reviews by Haynes et al. (1984) and Giovanelli & Haynes (1990). Here we focus only on the morphological dependence among normal objects.

Several quantities are generally used in analyzing the total HI content of galaxies: the total HI mass MHI, the hydrogen mass to luminosity ratio MHI/LB, and the HI surface density, sigmaHI. While both MHI/LB and sigmaHI have been used as the comparative measure of HI content, a residual dependence of the former ratio on LB exists, in the sense that higher luminosity galaxies have systematically lower values of MHI/LB. Numerous authors (e.g., Bottinelli et al. 1974) have shown that caution is necessary in using MHI/LB if the Malmquist bias might play a role. Furthermore, since LB includes contributions from both disk and bulge, and the HI is a disk property, the morphological dependence on MHI/LB is complicated. Finally, the fraction of the total mass in the form of HI can be examined via the ratio MHI/MT. Figures 4a-c show the results of our analysis of HI properties.

The total HI mass MHI is a scale parameter that is seen to vary over at least 4 orders of magnitude. It is well know that early-type systems - E and S0 galaxies - contain proportionately lower HI masses, and in fact show a much larger range in all measures of HI content, both MHI and sigmaHI relative to the later spirals. While some E's and S0's have HI contents similar to those of Sb-Sc spirals, others contain several orders of magnitude less HI. For this reason and because the HI within S0's is often located in an annulus exterior to the optical disk, van Driel and van Woerden (1991) and others have suggested that the HI gas has an external origin, the result of a tidal interaction or the infall of a dwarf companion. Bregman et al. (1992) propose that almost no true E's have detectable HI gas except for those few instances where, through the HI kinematics and distribution, infall is indicated. As evident in Figure 4c, the fractional HI mass increases systematically from E/S0 to Im.

Among the later-type spirals, sigmaHI is useful as an indirect probe of the effect of local environment on star formation potential. In the study of the HI deficiency of cluster galaxies relative to their counterparts in low density regions, Haynes et al. (1984) have defined the measure of HI content as the difference between the observed HI mass (in logarithmic units) and that expected for a galaxy of the same linear diameter and morphological type in a comparison sample of isolated objects. Specifically, the HI deficiency parameter is defined as

< DEF > = log [MHI (T, D)]0 - log [MHI (T, D)obs].

The use of the HI deficiency parameter has led numerous authors to conclude that spiral galaxies that pass through a hot X-ray intracluster medium are stripped of their HI gas. Warmels (1988a, b) and Cayatte et al. (1990) find that the HI disks of Virgo core galaxies are indeed shrunken with respect to their field counterparts or objects outside the core. Haynes & Giovanelli (1986) found a one-to-one correspondence between high HI deficiency and shrunken HI size. Although ram pressure stripping is the favored explanation for the HI deficiency, analysis of the the current data do not discriminate among the alternatives of ram pressure, conductive heat transport, turbulent viscosity or tidal effects (Magri et al. 1988). In other instances, slow but close, prograde tidal encounters can similarly remove the majority of a galaxy's interstellar HI. The effect of environment is discussed further in Section 4.

Next Contents Previous